We present a novel implementation of compressed suffix arrays exhibiting new tradeoffs between search time and space occupancy for a given text (or sequence) of n symbols over an alphabet Σ, where each symbol is encoded by lg |Σ | bits. We show that compressed suffix arrays use just nHh + O(n lg lg n / lg |Σ | n) bits, while retaining full text indexing functionalities, such as searching any pattern sequence of length m in O(m lg |Σ | + polylog(n)) time. The term Hh ≤ lg |Σ | denotes the hth-order empirical entropy of the text, which means that our index is nearly optimal in space apart from lower-order terms, achieving asymptotically the empirical entropy of the text (with a multiplicative constant 1). If the text is highly compressible so that Hh = o(1) and the alphabet size is small, we obtain a text index with o(m) search time that requires only o(n) bits. Further results and tradeoffs are reported in the paper. 1

Full-text indexes provide fast substring search over large text collections. A serious problem of these indexes has traditionally been their space consumption. A recent trend is to develop indexes that exploit the compressibility of the text, so that their size is a function of the compressed text length. This concept has evolved into self-indexes, which in addition contain enough information to reproduce any text portion, so they replace the text. The exciting possibility of an index that takes space close to that of the compressed text, replaces it, and in addition provides fast search over it, has triggered a wealth of activity and produced surprising results in a very short time, and radically changed the status of this area in less than five years. The most successful indexes nowadays are able to obtain almost optimal space and search time simultaneously. In this paper we present the main concepts underlying self-indexes. We explain the relationship between text entropy and regularities that show up in index structures and permit compressing them. Then we cover the most relevant self-indexes up to date, focusing on the essential aspects on how they exploit the text compressibility and how they solve efficiently various search problems. We aim at giving the theoretical background to understand and follow the developments in this area.

Abstract. Given a sequence S = s1s2... sn of integers smaller than r = O(polylog(n)), we show how S can be represented using nH0(S) + o(n) bits, so that we can know any sq, as well as answer rank and select queries on S, in constant time. H0(S) is the zero-order empirical entropy of S and nH0(S) provides an Information Theoretic lower bound to the bit storage of any sequence S via a fixed encoding of its symbols. This extends previous results on binary sequences, and improves previous results on general sequences where those queries are answered in O(log r) time. For larger r, we can still represent S in nH0(S) + o(n log r) bits and answer queries in O(log r / log log n) time. Another contribution of this paper is to show how to combine our compressed representation of integer sequences with an existing compression boosting technique to design compressed full-text indexes that scale well with the size of the input alphabet Σ. Namely, we design a variant of the FM-index that indexes a string T [1, n] within nHk(T) + o(n) bits of storage, where Hk(T) is the k-th order empirical entropy of T. This space bound holds simultaneously for all k ≤ α log |Σ | n, constant 0 < α < 1, and |Σ | = O(polylog(n)). This index counts the occurrences of an arbitrary pattern P [1, p] as a substring of T in O(p) time; it locates each pattern occurrence in O(log 1+ε n) time, for any constant 0 < ε < 1; and it reports a text substring of length ℓ in O(ℓ + log 1+ε n) time.

Abstract. We present a linear-time algorithm to compute the longest common prefix information in suffix arrays. As two applications of our algorithm, we show that our algorithm is crucial to the effective use of block-sorting compression, and we present a linear-time algorithm to simulate the bottom-up traversal of a suffix tree with a suffix array combined with the longest common prefix information. 1

Let a text of u characters over an alphabet of size be compressible to n symbols by the LZ78 or LZW algorithm. We show that it is possible to build a data structure based on the Ziv-Lempel trie that takes 4n log 2 n(1+o(1)) bits of space and reports the R occurrences of a pattern of length m in worst case time O(m log(m)+(m+R)log n).

In this paper we consider the problem of computing the suffix array of a text T [1, n]. This problem consists in sorting the suffixes of T in lexicographic order. The suffix array [16] (or pat array [9]) is a simple, easy to code, and elegant data structure used for several fundamental string matching problems involving both linguistic texts and biological data [4, 11]. Recently, the interest in this data structure has been revitalized by its use as a building block for three novel applications: (1) the Burrows-Wheeler compression algorithm [3], which is a provably [17] and practically [20] effective compression tool; (2) the construction of succinct [10, 19] and compressed [7, 8] indexes; the latter can store both the input text and its full-text index using roughly the same space used by traditional compressors for the text alone; and (3) algorithms for clustering and ranking the answers to user queries in web-search engines [22]. In all these applications the construction of the suffix array is the computational bottleneck both in time and space. This motivated our interest in designing yet another suffix array construction algorithm which is fast and "lightweight" in the sense that it uses small space...

We introduce two succinct data structures to solve various string problems. One is for storing the information of lcp, the longest common prefix, between suffixes in the suffix array, and the other is an improvement in the compressed suffix array which supports linear time counting queries for any pattern. The former occupies only 2n + o(n) bits for a text of length n for computing lcp between adjacent suffixes in lexicographic order in constant time, and 6n + o(n) bits between any two suffixes. No data structure in the literature attained linear size. The latter has size proportional to the text size and it is applicable to texts on any alphabet &Sigma; such that |&Sigma;| = log^O(1) n. These space-economical data structures are useful in processing huge amounts of text data.

A succinct full-text self-index is a data structure built on a text T = t1t2...tn, which takes little space (ideally close to that of the compressed text), permits efficient search for the occurrences of a pattern P = p1p2... pm in T, and is able to reproduce any text substring, so the self-index replaces the text. Several remarkable self-indexes have been developed in recent years. Many of those take space proportional to nH0 or nHk bits, where Hk is the kth order empirical entropy of T. The time to count how many times does P occur in T ranges from O(m) to O(m log n). In this paper we present a new self-index, called RLFM index for “run-length FM-index”, that counts the occurrences of P in T in O(m) time when the alphabet size is σ = O(polylog(n)). The RLFM index requires nHk log σ + O(n) bits of space, for any k ≤ α log σ n and constant 0 < α < 1. Previous indexes that achieve O(m) counting time either require more than nH0 bits of space or require that σ = O(1). We also show that the RLFM index can be enhanced to locate occurrences in the text and display text substrings in time independent of σ. In addition, we prove a close relationship between the kth order entropy of the text and some regularities that show up in their suffix arrays and in the Burrows-Wheeler transform of T. This relationship is of independent interest and permits bounding the space occupancy of the RLFM index, as well as that of other existing compressed indexes. Finally, we present some practical considerations in order to implement the RLFM index, obtaining two implementations with different space-time tradeoffs. We empirically compare our indexes against the best existing implementations and show that they are practical and competitive against those. 1

In this work, we introduce the new problem of finding time series discords. Time series discords are subsequences of a longer time series that are maximally different to all the rest of the time series subsequences. They thus capture the sense of the most unusual subsequence within a time series. Time series discords have many uses for data mining, including improving the quality of clustering, data cleaning, summarization, and anomaly detection. As we will show, discords are particularly attractive as anomaly detectors because they only require one intuitive parameter (the length of the subsequence) unlike most anomaly detection algorithms that typically require many parameters. We evaluate our work with a comprehensive set of experiments. In particular, we demonstrate the utility of discords with objective experiments on domains as diverse as Space Shuttle telemetry monitoring, medicine, surveillance, and industry, and we demonstrate the effectiveness of our discord discovery algorithm with more than one million experiments, on 82 different datasets from diverse domains.

We report on a new and improved version of high-order entropy-compressed suffix arrays, which has theoretical performance guarantees similar to those in our earlier work [16], yet represents an improvement in practice. Our experiments indicate that the resulting text index offers state-of-the-art compression. In particular, we require roughly 20 % of the original text size—without requiring a separate instance of the text—and support fast and powerful searches. To our knowledge, this is the best known method in terms of space for fast searching. 1